montane valley grasslands are highly resistant to summer...

J Veg Sci 2018291017ndash1028 wileyonlinelibrarycomjournaljvs emsp|emsp1017

Journal of Vegetation Science

copy 2018 International Association for Vegetation Science

1emsp |emspINTRODUC TION

Changes in ecological communities are largely driven by distur-bance events that take place across a variety of temporal and spatial scales (Pickett amp White 1985) These disturbances result in a mosaic of patch types that differ in age thus creating a source of spatial

and temporal heterogeneity This is especially the case in grassland ecosystems where disturbed patches often support a species as-semblage that is for a period of time different from that of pre- disturbed conditions (Collins amp Smith 2006) These patches vary in resource availability species composition vegetation structure and ecosystem processes within a region (Milchunas amp Lauenroth 1993 White amp Jentsch 2001)

Anthropogenic disturbances such as livestock grazing (VanAuken 2009) development (York 2011) and fire suppression (Allen 1989

Received5June2018emsp |emsp Revised25July2018emsp |emsp Accepted2October2018DOI101111jvs12690

R E S E A R C H A R T I C L E

Montane valley grasslands are highly resistant to summer wildfire

Martina M Suazo12 emsp|emspScott L Collins2emsp|emspRobert R Parmenter12emsp|emspEsteban Muldavin3

Nomenclature USDA Plants Database (httpplantsusdagovjava accessed 5 Mar 2018)

1National Park Service Valles Caldera National Preserve Jemez Springs New Mexico2Department of Biology University of New Mexico Albuquerque New Mexico3Department of Biology Natural Heritage New Mexico University of New Mexico Albuquerque New Mexico

CorrespondenceMartina M Suazo National Park Service Valles Caldera National Preserve Jemez Springs NMEmail martina_suazonpsgov

Funding informationThe National Park Service The Southwest Jemez Mountains Collaborative Forest Landscape Restoration Program

Co-ordinating Editor David Ward

AbstractAims Understanding the ecological role of fire in shaping plant communities in fire- prone ecosystems is needed for ecosystem restoration preservation and manage-ment We investigated the effects of wildfire on plant community structure and the biotic and abiotic factors that are most influential in stabilizing andor driving change before and after burning in high- elevation montane grasslands dominated by C3 spe-cies receiving a mean annual precipitation of 545 mmLocation Valles Caldera National Preserve Jemez Mountains New Mexico USAMethods Long- term data (10 year pre- fire 5 year postfire) on plant community composition of nine burned and seven unburned grassland sites were used to deter-mine the response of montane grasslands to a summer wildfire We used multivariate and univariate analyses to determine changes in plant community composition and structure pre- and post- wildfire on burned and unburned grasslandsResults The montane grassland community consisted of 155 plant taxa including 44 perennial grasses one annual grass 92 perennial forbs and nine annual forbs We found that these fire- adapted plant communities were highly resilient to fire fire had no significant effects on composition or structure beyond the normal range of inter- annualvariabilityInsteadseasonalprecipitationhadthelargestinfluenceonplantcommunity dynamics over time with lower average plant canopy height and diversity during drought periodsConclusionOurresultsshowthatfirehasverylimitedeffectsonthecompositionand structure of these C3-dominatedmontanegrasslandsOur results supportef-forts by land managers to reintroduce fire on these historically burned landscapes

K E Y W O R D S

C3 grasses community dynamics montane valley grassland resilience wildfire

1018emsp |emsp emspenspJournal of Vegetation Science

SUAZO et Al

Coop amp Givnish 2007a VanAuken 2009) have all contributed to the decline of grassland ecosystems in western North America Fire is an essential ecological component for maintaining the health and proper functioning of forests across the southwestern United States by maintaining community structure ecosystem functioning and species diversity via fuel removal and the release of organically stored nutrients (Dewar 2011) Because montane grasslands are in-terspersed throughout mountain ranges within forest stands there can be important fire regime interactions at a landscape scale that have implications for impacts in more localized habitats For exam-ple in montane grasslands of the Jemez Mountains in north- central New Mexico frequent fires (fire return intervals of lt 10 year) typ-ically move along forestndashgrassland ecotones killing seedling trees encroaching into grasslands (Conver Falk Yool amp Parmenter 2018) Fire has been the single most important factor in preventing forest encroachment and maintaining the integrity of grassland communi-ties however as a result of anthropogenic- induced fire suppression nearly 18 of grassland in this area has been lost to tree encroach-ment between 1935 and 1996 (Coop amp Givnish 2007a)

Most grassland studies of fire impacts focus on warm- season C4- dominated systems at relatively lower elevation with lower av-erage annual precipitation Studies have found that fire favours C4 grasses (Collins amp Calabrese 2012 Tix amp Charvat 2005) and that the expansion of C4 grasses in the late Miocene was driven by frequent fire (Scheiter et al 2012) Therefore it is generally thought that C4 species are better adapted to fire than cool- season C3 species (RipleyDonaldOsborneAbrahamampMartin2010)Consequentlymost studies on fire impacts to grasslands are conducted in C4- dominated grasslands thereby leaving the more mesic C3 grassland response to fire less well documented Also lacking have been the long- term studies of wildfire impacts that include knowledge of pre- fire conditions and postfire succession beyond the first two or three growing seasons following fire Furthermore the effects of fire on species in mountain grasslands are often studied in forest understo-ries rather than open grassland ecosystems

In this context we investigated the effects of an extensivesummer wildfire on plant community structure in the primarily C3- dominated montane valley grasslands of the Valles Caldera National Preserve located in the centre of the Jemez Mountains in north-ern New Mexico USA Using an established long- term pre- and post- wildfire grassland sampling array we asked (a) how did these montane grasslands respond in terms of plant species composition abundance and spatial distribution before and after the wildfire and (b) what biotic and abiotic factors are most influential in stabilizing andor driving change before and after burning

2emsp |emspMETHODS

21emsp|emspSite description

The Valles Caldera National Preserve (Preserve) is located in the Jemez Mountains of north- central New Mexico USA (Supporting

InformationAppendixS1)andconsistsofhigh-elevationforestandgrassland systems that lie at the head of the Jemez River watershed which flows into the Rio Grande Valley The 36017 ha Preserve is centred on a volcanic caldera approximately 24 km in diameter that erupted 12 million years ago (Goff 2009) and comprises large areas of low- lying mountain meadows and riparian vegetation with sloped mountain valley grasslands bordered by several forested lava domes ranging in elevation from 2590 to 3505 m An estimated 10522 ha of montane grassland occurs in the Preserve dominated mostly by C3 grasses that are typical of Southern Rocky Mountain MontanendashSubalpine Grassland per the US National Classification (httpusnvcorg) These include perennial bunch grasses such as Festuca idahoensis F arizonica F thurberi Danthonia parryi and Poa prat-ensis along with a relatively cold- tolerant C4 species Muhlenbergia montana which commonly occurs in subalpine grasslands across the southwestern US (Brown 1994) The grasslands of the Preserve have formed on prehistoric lake beds that are between 70000 and 500000+ years old whose soils consist primarily of Mollisols that have developed from older quaternary alluvial fan (Qf) deposits from the surrounding domes (Muldavin 2003)

Prior to 1900 fires occurred on average every 16 year some-where on Preserve grasslands with widespread fires occurring at decadal intervals and smaller low intensity fires particularly in valley grasslandsoccurringevery27ndash101year(Dewar2011)Inadditionthe Preserve has undergone extensive livestock grazing during the 19th and 20th centuries but currently supports only a relatively small (lt800 head) herd of cattle and a population of approximately 2000 elk (Cervus canadensis) native to the region (cattle have grazed all study sites to a limited degree both pre- and postfire) Mean an-nualprecipitationderived from the2004ndash2015wateryears (OctndashSept) is 545 mm 60 of this precipitation is produced by monsoonal rainfall between the months of June and September and winter pre-cipitation is primarily in the form of snowpack (data from Western Regional Climate Center httpswrccdrieduvallescaldera)

In summer 2011 the Las Conchas Wildfire burned through63131 ha of forest and grassland vegetation and at the time was the largest wildfire in recorded New Mexico history Fire intensity varied from stand replacement in the forest to low- intensity ground fire in the open grasslands The grassland sites in this study burned over the course of 5 days from 27 June to 1 July with an average wind speed of 2 ms and gusts with a maximum of 171 ms There was 36 average relative humidity with a 103 minimum and 892 maximum Average air temperature for this time period was 180degC 36degC minimum and 283degC maximum

22emsp|emspField data collection

A rangeland monitoring programme was established in the Preserve in 2001 consisting of 44 vegetation monitoring sites (Barnes 2003) These sites were stratified by relatively homogenous repeating ecotypes which were identified by soil type land form community composition management history and future management poten-tial Three ecotypes occur across the Preserve Mountain Meadow

emspensp emsp | emsp1019Journal of Vegetation Science

SUAZO et Al

Mountain Valley and Grazeable Woodland (Barnes 2003) For our analysis of fire effects on montane grassland communities we used 16 monitoring sites from the Mountain Valley (MV) ecotype that were representative of Southern Rocky Mountain Montane Grasslands with respect to species dominance and site conditions

The MV ecotype is located on the upper slope margins of the valleys which consists of open grasslands dominated by native bunch grasses particularly the C3 grasses Festuca arizonica and Danthonia parryi (Barnes 2003) Downslope the grasslands grade to wet meadows and herbaceous wetlands and upslope to mixed coniferous forests The MV ecotype is composed of two different soil types Tranquilar- Jarmillo complex (series 302) characterized by a surface horizon of silt clay loamloam to 33 and 91 cm in depth respectively Cosey- Jarmillo association (series 304) with a silt loam to 38 cm (NCSS 1987)

Of16availableMVsitesninewerecompletelyburnedin201110 years after their establishment and these serve as the treat-ment group The other seven sites were not burned over the entire 15yearsofrecordandserveasthereferenceorcontrolgroupOuron- site fire observations were that the 2011 fire was low intensity and moved rapidly and more or less uniformly across the surface of the sites (Figure 1)

Each monitoring site consisted of three 100- m transect lines radiating outward from a central location at 0ordm 120ordm and 240ordm The line- point intercept sampling method was used to identify and

measure plant species present at each meter for a total of 100 points per line and 300 points per site Measurements were taken with the use of a thin steel rod or dowel 12 m in length by 1 cm in diameter The species of every live plant touching the rod or intersecting the vertical line drawn by the rod from the top of the plant canopy down to the ground surface was recorded to species level in most cases The surface substrate touched by the base of the rod at ground level was also recorded and used to deter-minebasallittercoverageIfonespeciesoccurredmorethanonceat a particular point only its highest appearance was recorded Canopy height measurements estimated to the nearest centime-ter were recorded as the height of the point at which the tallest plant intersected the sampling rod Where there was no live can-opy the height was recorded as 0 cm Two photos were taken of each transect line with the measuring tape present on the ground for a total of at least six photographs per site per year Photo mon-itoring and data collection were conducted annually during the growing season (JunndashSept) Long- term measurement of grassland species composition at these sites was conducted from 2001 until present although not all sites were sampled in all years (Table 1) Raw data and a complete species list are presented in Supporting InformationAppendicesS2andS3respectively

Annual peak standing crop biomass was measured by clipping all herbaceous vegetation at ground level at each sampling site using four replicate 025 m2 rings randomly placed inside of ungulate

F IGURE 1emspTime series photographs for paired burn and control sites (a) 2005 pre- burn for burned site MV06 (b) 2011 immediate post- burn for burned site MV06 (c) 2012 post- burn for burned site MV06 (d) 2005 pre- burn for control site MV09 (e) 2011 immediate post- burn for control site MV09 (f) 2012 post- burn for control site MV09 [Colour figure can be viewed at wileyonlinelibrarycom]

(a) (d)

(b) (e)

(c) (f)


SUAZO et Al

TABLE 1emsp

Site

cha

ract

eris

tics

for a

ll m

onito

ring

plot

s

Site

Trea

tmen

tTo

tal y

ears

gr

azed

Year

s gra

zed

pre-

fire

Year

s gra

zed

post

- fire

Year

s NO

T sa

mpl

edSo

il se

riesa

Asp

ect (

deg)Sl

ope

El

evat

ion

(m)

Dom

inan

t gra

ssb

MV

02BU

RN6

60

130

221

12

260

7FESX1MUMOPOPR

MV

03BU

RN6

60

330

818

15

261

1FESX1MUMOPOPR

MV

04BU

RN7

34

130

215

34

262

3DAPAPOPRFESX1

MV

05BU

RN6

24

230

468

52

734

DAPAPOPRFESX1

MV

06BU

RN5

50

130

245

32

656

DA

PA F

ESX1

MV

07BU

RN6

60

530

432

88

269

2D

APA

FES

X1

MV

08BU

RN5

50

130

419

88

270

2D

APA

FES

X1

MV

09CONTROL

66

01

302

168

22

585

FESX1MUMO

MV1

0CONTROL

66

02

302

245

42

559

FESX1MUMO

MV1

1CONTROL

116

53

302

269

112

574

FESX1POPR

MV1

2CONTROL

116

52

304

626

268

5DAPAPOPRFESX1

MV1

3CONTROL

116

53

304

274

266

2DAPAPOPRFESX1

MV1

5BU

RN12

75

430

467

82

650

DAPAPOPRFESX1

MV1

7CONTROL

138

52

302

123

32

608

POPR

MV1

8BU

RN12

75

530

250

72

670

DAPAPOPRFESX1

MV

20CONTROL

116

54

304

365

261

4DAPAPOPR

a Soil

Serie

s ta

ken

from

USD

A N

RCS

Soil

Surv

ey o

f San

dova

l Cou

nty

Are

a N

ew M

exic

o b D

omin

ant p

lant

spe

cies

by

abun

danc

e D

APA

Dan

thon

ia p

arry

i FE

SX1

Fes

tuca

ariz

onic

a an

d Fe

stuc

a id

ahoe

nsis

consolidatedMUMOM

uhle

nber

gia

mon

tanaPOPRP

oa p

rate

nsis


SUAZO et Al

grazing exclosures approximately 2 times 2 m in size Biomass was dried at 60degC for 48 hr and weighed to the nearest gram

Some closely related species present at these sites have similar morphology and were identified inconsistently by different sampling teams over time To address this problem the following species were combined at the genus level Antennaria parvifolia Antennaria rosea Antennaria rosulata Festuca arizonica Festuca idahoensis Bromus anom-alus Bromus porteri and multiple sedge species

23emsp|emspStatistical analysis

Assessments within and between burn and control sites 10 years be-fore and 5 years after the 2011 fire were used to determine spatial and temporal response of grassland communities to fire Abundance data for all plant species encountered at each survey site for all years sampled were aggregated and transformed to the 4th root in order to achieve the highest level of normality possible We used NMDS with a Bray- Curtis similarity index to display spatial variability in species composition among sites and treatments post- burn trajectories for each site and distribution and abundance of the four most common grassesWethenranPERMANOVAwithaBray-Curtissimilarityindexto make pairwise comparisons between the following pre- fire vs postfire among burned sites pre- fire vs postfire among unburned sites and post-fireburnedvspost-fireunburnedsitesWeusedSIMPERtoclas-sify each monitoring site based on the top four most abundant species present across all sites andSIMPERwasalsoused to identifywhichspecies contributed most to compositional differences temporally and spatially The species with the largest contribution to compositional dif-ferences were determined to be those in the top 15 cumulative con-tribution range All multivariate analyses were conducted using Primer 6(PRIMER-EPlymouthUK)

Repeatedmeasures ANOVAwas used to test for differences inplant canopy height species richness species evenness and grass and

forb abundance between the burn and control groups for all years sam-pledusingthetreatmentandyearinteractiontermInordertoreducethe influence of extraneous variation on the response we used a ran-domized block design (RBD) with each site treated as a separate block and sampling years treated as replicates for two treatment types ndash be-fore (2001ndash2010) and after (2011ndash2015) fire in order to identify tempo-ral changes in plant canopy height species richness species evenness and grass and forb abundance Lastly resilience was measured using ln(BiomassyearTotal Average Biomass2002ndash2010 Tilman amp Downing 1994) with significance testing using one sample t- test between each postfire year and the 10- year (pre- fire) average in annual standing biomass

To determine the influence of abiotic drivers on plant commu-nity structure we used simple linear regression to correlate mon-soon season precipitation accumulation over the months of June and July prior to the sampling with plant canopy height species richness species evenness and abundance of grass and forb functional types for all sites combined Precipitation data were acquired from four different meteorological stations located throughout the Preserve Each site was paired with its nearest weather station to most accu-rately represent the local weather conditions Furthermore percent-age ground covered by litter a factor known to affect community structure of herbaceous communities (Facelli amp Pickett 1991) was correlated to species evenness over time for all sites combined using Spearman ranked correlation in order to identify the role of this bi-otic factor in community change after fire

3emsp |emspRESULTS

We found a total of 155 plant taxa across all sites and years with grasses accounting for 22 of total species richness (44 perennial one annual) and half of the total canopy cover while forbs account for 62 of species richness (92 perennial nine annual) but only 27 of total

F IGURE 2emspNMDS bubble plots show spatial distribution of the four most common species across all sites between the two time periods (before and after fire) arrows in the Festuca spp illustrate direction of temporal trajectory for all four panels Bubble size illustrates relative species abundance therefore scaling is not standardized across all four panels


SUAZO et Al

canopycover(SupportingInformationAppendixS3)Sitesweredomi-nated by six combinations of four abundant grass taxa Danthonia par-ryi Festuca spp Muhlenbergia montana and Poa pratensis (Table 1) We found these species to be present at every site and also to be the most abundant overall however their abundance and distribution varied widelyacrosssites(Figure2)BasedonPERMANOVAthisvariationinabundance drives significance between site comparisons where species

composition among all 16 sites was significantly different prior to the fire therefore these sites were not true replicates as initially assumed Each site consists of the same suite of species but with frequencies that fluctuate enough to create a different community configuration and re-sult in significant spatial difference prior to wildfire Clustering of sites by species dominance occurs to some extent where sites to the right in the NMDS lack D parryi but have high levels of M montana and Festuca

TABLE 2emspPERMANOVAandSIMPERresultsofpair-wisecomparisonsforplantcommunityassemblagesinburnedandcontroltreatmentgroups before and after fire for 2010 and 2011 only

A Community pairwise comparisons

Pre (2010)ndashPost (2011)

PERMANOVA PERDISPa

n t p Permutations t p Permutations

Pre- fire Post- fireBURN 16 9 17 0002 999 0 09 999

Pre-firePost-fireCONTROL 16 7 13 01 996 0 07 999

Post- fireBURN Post- fireCONTROL

9 7 12 01 963 0 08 999

B SIMPER (Significant contributors)

Pre- fire post- fireBURN species

Pre- fire AvAbund

Post- fireBURN AvAbund AvDiss DissSD Contrib Cum

Danthonia parryi 236 259 147 109 423 423

Agrostis scabra 12 0 112 157 323 747

Trifolium repens 14 045 108 151 312 1058

Poa fendleriana 157 085 108 133 311 137

p-value 005 = Abundance values are based on data transformation to the 4th root Comparisons with significance are highlightedaPERDISPisapermutationaldispersiontestietestofvariabilityorspatialspreadbetweengroups

TABLE 3emspPERMANOVAandSIMPERresultsofpair-wisecomparisonsforplantcommunityassemblagesinburnedandcontroltreatmentgroups before and after fire for years 2010 and 2012 only

A Community pairwise comparisons

Pre(2010)- Post(2012)

PERMANOVA PERDISPa





7 7 07 09 744 13 02 999

B SIMPER (significant contributors)

Pre- fire Post- fireCONTROL species

Pre- fire AvAbund

Post- fireCONTROL AvAbund AvDiss DissSD Contrib Cum



Bouteloua gracilis 044 114 106 117 302 1093

Agrostis scabra 12 034 102 141 289 1382



SUAZO et Al

spp (Figure 2) Sites to the upper left of the NMDS have low frequency of P pratensis and abundant D parryi and are also geographically close together Sites in the lower left of Figure 2 are all dominated by D par-ryi P pratensis and Festuca spp with similar relative abundances across sites cattle have grazed all of these sites during the 5 year following fire (Table 1)

Immediate post-burn effectswere tested by comparing commu-nitycompositionbetweenyears2010and2011usingPERMANOVASignificant differences occurred between pre- fire and the postfire

burn treatment group but not for the unburned sites (Table 2A) Based onSIMPERthedifferencesbetweenpre-andpost-burnweredrivenby a small increase in Danthonia parryi and declines in the abundance of Agrostis scabra Trifolium repens and Poa fendleriana (Table 2B) Based on our comparisons for years 2010 and 2012 however these dif-ferences were no longer significant due primarily to the recovery of A scabra We also found significant differences between the pre- fire and postfire unburned control group indicating an effect of factors other than fire (Table 3A) This dissimilarity was driven by most of the

TABLE 4emspPERMANOVAandSIMPERresultsofpair-wisecomparisonsforplantcommunityassemblagesinburnedandcontroltreatmentgroups before and after fire for 15- year study period (2001ndash2015)


Pre(2001ndash2010)- Post(2011ndash2015)

PERMANOVA PERDISPa





40 28 16 001 999 2 0009 999


Pre- fire Post- fireBURN species

Pre- fire AvAbund



Elymus elymoidies 115 137 102 137 281 704

Poa fendleriana 143 092 099 128 272 976

Bryophyte species 125 08 097 128 266 1242

Taraxacum officinale 127 155 093 122 257 1499

SIMPER Significany contributors


Pre- fire AvAbund




Poa fendleriana 143 108 096 123 259 96



SIMPER Significance contributors

Post- fireBURN Post- fireCONTROL Species

Post- fireBURN AvAbund





Poa fendleriana 092 108 089 117 24 1181

Symphyotricum ascendens 09 084 084 125 228 1409



SUAZO et Al

same species that contributed to the difference between the pre- fire andpost-fireburngroupsIngeneraltherewasarelativelylargede-cline in the average abundance of D parryi T repens and A scabra in addition to an average increase in Bouteloua gracilis in 2012 compared to 2010 (Table 3B)

When all 16 sampling sites for the entire 15- year study period were grouped according to treatment all pairwise comparisons (pre- firendashpost- fire burnedndashunburned) were found to be significantly different (Table4A)BasedonSIMPERthelargestcontributorstothedifferencebetween the pre- fire and postfire burn groups were D parryi Elymus elymoides and Taraxacum officinale which all increased in abundance on average along with declines in Poa fendleriana and non- vascular bryophyte species (Table 4B) The significance between pre- fire and the postfire control group was explained by an average decrease in D parryi P fendleriana and bryophytes as well as an average increase in B gracilis and E elymoidesOveralltemporaldifferenceswithinsiteswere much smaller than the spatial differences among sites but there was no consistent direction or degree of dispersion of samples in the NMDS after fire within the two treatment groups (Figure 3) Trends within each treatment group were not apparent and both treatment groups behaved similarly following the fire Also despite no clear clus-tering of control and burn groups either before or after fire they were significantly different after fire (Table 4A)

Inadditionpre-andpost-firecomparisonsweremadeovertheentire15-yearstudyperiodforeachsiteindividuallyBasedonPERMANOVAnot all sites differed significantly between the two time periods Eight of the 16 sites were significantly different postfire and these consisted of five burn and three control sites The other half was not significantly differentpost-fireandconsistedoffourburnandfourcontrolsitesInthese temporal comparisons the only detectable change in composition from fire was the loss of Symphyotrichum ascendens and Deschampsia caespitosa at two sites (MV7 and MV18 respectively) in 2011 and they have not reappeared since the fire These two species were relatively

rare with average cover values less than 2 S ascendens was present at all other sites but Deschampsia was not Rare species (in terms of fre-quency) such as bryophytes A scabra P fendleriana Vicia americana and Symphyotrichum contributed most to the differences seen postfire for each site individually but in general the species with the largest contribution to dissimilarity varied across sites and was not consistent withintreatmentgroupOfthespeciesthatcontributedtothetop15cumulative abundance some increased (V americana) while others de-creased (A scabra P fendleriana S ascendens) after 2011 but again some of these declined in the control sites as well so a clear effect of fire was not detectable

Annual standing crop biomass immediately after the fire in 2011 was significantly lower than the pre- fire average (p = 00001) yet biomass fully recovered to pre- fire levels by 2012 Standing biomass for each postfire year from 2012 to 2014 did not differ from the 10- year pre- fire average however biomass levels for 2015 were sig-nificantly (p = 0002) higher than the pre- fire average (Figure 4) This temporal trend tracks average precipitation accumulation for the months of June and July as well as for the full water year (Supporting InformationAppendixS4)

BasedonrepeatedmeasuresANOVAnosignificantdifferencesoccurred between the burn and control treatment groups in any year for plant canopy height species richness species evenness grass and forb abundance (Table 5) The randomized block design (RBD) tests found that all variables other than species richness were sig-nificantly higher in both burned and unburned areas after the fire (Table 5)

All five community structure variables were positively correlated with summer monsoonal precipitation but not winter or annual to-tals (Figure 5) This indicates that temporal patterns in plant com-munity structure are strongly driven by climate overshadowing any effect of burning for all years even after the fire Litter as a biotic influence on community trends was negatively correlated with spe-cies evenness (r2=minus067p lt 0001)

4emsp |emspDISCUSSION

Fire is a natural disturbance that can be used as a management tool for maintaining healthy ecosystems particularly for those that have

F IGURE 3emspNMDS of the pre- and post- burn cluster centroids for all 16 sites arrows illustrate direction of temporal trajectory Circle=BurnSquare=ControlOpensymbols=Pre-FireFilledsymbols = Post- Fire

F IGURE 4emspAverage annual standing crop biomass in gm2 including long term pre- fire average (2001ndash2010) and all postfire years


SUAZO et Al

evolved with fire (Wright amp Bailey 1980) The potential negative im-pacts of fire include the loss of herbaceous cover which leads to water run- off and erosion consequently impacting water quality and water-shed function (Dahm Candelaria- Ley Reale Reale amp Van Horn 2015)

Fire may increase opportunities for invasion or colonization of unde-siredspecies (HunterOmiMartinsonampChong2006)andchangeplant community composition in a way that impacts habitat type and quality leading to altered trophic structure and dynamics (Ford amp McPherson 1996) However despite the intensity and size of the Las Conchas wildfire the burn had few impacts on the montane val-ley grasslands of the Valles Caldera National Preserve There were no differences in plant canopy height species richness species evenness and grass and forb abundance between the burn and control groups for any year including the year immediately after the fire Pre- fire bio-mass conditions were reached by the second growing season after fire and no new exotic species were detected naturalized species such as P pratensis followed the same postfire temporal recovery trend in abundance as native species These results lead us to suggest that this montane grassland system is highly resilient to wildfire

We acknowledge the inherent variability in different wildfire events and note that our results are technically limited to a single wildfire event Natural wildfires (unlike prescribed burns) are rarely replicatedduetounknownfire-returnprobabilitiesInourstudytheLas Conchas fire burned for five consecutive days and we sampled multiple sites across the landscape that had burned on different days Hence the Las Conchas fire comprised a series of daily fire events (each one slightly different from the others) that could serve as replicated fire treatments These circumstances have the advan-tage of controlling for sitehabitat characteristics while maintaining

TABLE 5emspResultsofrepeatedmeasuresANOVAandrandomizedblock design for five plant community variables

df Sum Sq F- value p- value

RepeatedmeasuresANOVA Treatment Year

Plant height 1 17 0301 059

Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037

Random block design BeforeAfter fire

Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =

F IGURE 5emspRelationship between cumulative seasonal precipitation (June plus July) and plant canopy height species richness species evenness and grass and forb abundance


SUAZO et Al

approximately the same weather conditions between each fire eventIngeneralfirestudiesareconsideredassingleeventswithsite- and time- specificity and the statistical ldquoreplicationrdquo for infer-ence comes with multiple studies in similar habitats over many years and wildfire events The Las Conchas fire burned on the Preserve during a period of near- record drought representing an extreme fire event As such we consider our results of minimal fire impact on these montane grasslands to be indicative of the highly resilient na-ture of these grassland communities

The Preserversquos grasslands follow a common trend observed in other grassland systems where a small number of dominant species account for the majority of herbaceous cover while the majority of species are relatively rare (Collins amp Glenn 1991) The only detect-able change in composition from fire was the loss of S ascendens and D caespitosa following the burn Because the relative abundance of S ascendens is low it may still be present in the community but simply has not been documented since 2011 D caespitosa typically prefers moist habitats such as wet meadows and can tolerate all but the most severe fires (Debenedetti amp Parsons 1984) A decline in population may suggest a drying out of habitat resulting in condi-tions unsuitable for persistence However D caespitosa decreases with cattle grazing which may explain its relatively low abundance and subsequent inability to recover after fire under continued graz-ing pressure (Mueggler amp Stewart 1980)

All plant community analyses showed that the control group fol-lowed the same trends as the burn group after fire Fluctuations in community structure were not driven by fire nor were they due to postfire succession but rather they responded to other abiotic and biotic factors The species with the greatest contributions to signifi-cance between pre- fire and both postfire treatment groups for the full study period were similar suggesting that factors driving tem-poral change were consistent between the two treatments there-fore eliminating fire alone as a significant influence on community change Similarly a synthesis on fire effects in C4- dominated North American semiarid grasslands receiving lt600 mm mean annual pre-cipitation found a neutral to negative response in above- ground net primary productivity after fire regardless of season of fire grazing historyandmeanoractualprecipitation(ScheintaubDernerKellyampKnapp2009) Furthermore variability in above-groundnetpri-mary productivity (ANPP) which was similar to that of plant com-munity composition cover and diversity was not attributed to any particular site or fire characteristic

Significant positive correlations between precipitation and plant community characteristics showed that climatic variables had a greater influence on community structure in both treatment types than fire Similarly responses of ANPP in C3- dominated semiarid rangelands to precipitation in the northern mixed prairie of Montana were greater than those to fire grazing or the interaction of fire and grazing (Vermeire Crowder amp Wester 2014) These findings high-light the influence of water availability on these grassland systems and its role in shaping these plant communities relative to fire

Inadditionspeciesevennessshowedasignificantnegativecor-relation with ground cover of dead plant material across all sites

Litter accumulation can suppress plant growth therefore removal of dead plant material can increase germination by making resources such as sunlight more readily available (Foster amp Gross 1998 Scheintaub et al 2009 Xiong amp Nilsson 1999) However control sites that did not undergo burning also exhibited this same response which is likely due to litter reduction from cattle grazing given that five of the seven control sites were grazed in the 5 years following the2011fireOveralltheburnsitesinourstudywerenotgrazedbycattle to the same degree as the controls (Table 1) As a result litter removal whether by fire or grazing coupled with precipitation trends resulted in similar postfire responses across these grassland sites A conceptual model of grazing fire and climate in grasslands showed that annual fluctuations in ANPP increased as mean annual rainfall decreased and that 50 of inter- annual variability in biomass pro-duction resulted from deviation from mean precipitation in a given year(OesterheldLoretiSemmartinampParuelo1999)Ourfindingshere support this model where burning had little effect on compo-sition grazed and ungrazed sites were not consistently different postfire and the main driver of change in community structure and composition was monsoon season precipitation

Several studies have shown fire to favor expansion of C4 species (Collins amp Calabrese 2012 Scheiter et al 2012 Tix amp Charvat 2005) and have also found a C4 subspecies to be better adapted to fire than its C3 counterpart (Ripley et al 2010) Still uncertainty exists in determining if fire- adapted plant species are so as a result of their photosynthetic pathway or as a result of ad-aptations to a specific fire regime Results from an experimental fire in the northern Chihuahuan Desert in New Mexico showed that plant demography was highly impacted by fire (Parmenter 2008) of five abundant C4 grass species three were significantly negatively impacted by fireOverall in our study areawe showthat C3 plant species from open grasslands are adapted to wildfire and document a relatively high level of resistance and resilience There was no significant decline in average plant canopy height species diversity or abundance by functional group species even-ness was actually shown to increase after burning indicating higher resiliencetofirethanindesertgrasslandsItisunclearhoweverifthese grasslands are representative in terms of resilience of other high- elevation grasslands that evolved with fire or if the unique en-vironmental conditions of the caldera formation create a relatively higher resistance to fire

Fire management activities on the Preserve are focused on reducing the risk of landscape- level high- severity forest fires Because the Preserversquos landscape is a mosaic of forest stands and intervening valley grasslands (Figure 1) natural and man-aged fires tend to burn both forest and grassland habitats Ourresults support land and fire management efforts to maintain the use of fire as a tool in ecological restoration and preservation by providing evidence that any risk of adverse or unwanted impacts on plant community composition and function for high- elevation fire- adapted grasslands is minimal As in C4 grasslands (Collins amp Calabrese 2012) burning in the Preserve sustained diversity Furthermore maintaining fire frequency or fire return interval is


SUAZO et Al

crucial for maintaining ecosystem function because these grass-landsareadaptedtoaparticularfireregime(Dewar2011KeeleyPausas Rundel BondampBradstock 2011) In addition timingoffire is a crucial factor influencing the subsequent recovery of a plant community Cool season fires have minimal effects as herba-ceous plants are dormant For the Preserve natural wildfires typi-cally ignite during the driest part of the year (June) just prior to the start of the monsoon season (Dewar 2011)

It isunknownif thepotential formorefrequenthigh-intensityfire under climate warming will alter these low- intensity fire- adapted ecosystems in terms of community composition and eco-logical function Indeed it is possible that as climate continues towarm we may see that future high- intensity forest fires may actually expand montane grasslands What is not known is how these newly created grasslands will respond to continued burning and at what level of intensity future burning will reach given the amount of sur-face fuel remaining However in the context of this study a signif-icant impact to these plant communities from wildfire burning was not detected outside a normal range of variability therefore demon-stratingarelativelyhighlevelofresistancetofireInconclusionourresults demonstrate that these primarily C3- dominated montane grasslands are highly resistant to fire Therefore the use of fire as a management tool will help maintain the health and resilience of this landscape as well as reduce the loss of grassland to shrub and tree encroachment driven by fire suppression (Allen 1989 Coop ampGivnish2007a) Inadditionnatural systems that areextremelyresistant to environmental extremes should be preserved and pro-tected from anthropogenic development activities as they may be able to naturally resist climate change better than other less resis-tantresilient ecosystems

ACKNOWLEDG EMENTS

We thank the Valles Caldera National Preserve under both the United States Department of Agriculturersquos Valles Caldera Trust and theDepartmentof InteriorrsquosNationalParkService forsupportingthe long- term grassland monitoring programme botanist Will Barnes for the establishment of this monitoring programme as well as the numerous field botanists who have assisted with the data collection over the many years of this study Financial support for the project was provided by the National Park Service and the Southwest Jemez Mountains Collaborative Forest Landscape Restoration Program Additional field assistance for biomass sampling was provided by nu-merous citizen volunteers and the Sierra Club

ORCID

Martina M Suazo httporcidorg0000-0002-9312-2570

R E FE R E N C E S

Allen C D (1989) Changes in the landscape of the Jemez Mountains New Mexico PhD thesis University of California Berkeley Berkeley CA

Barnes W S (2003) 2002 Rangeland monitoring report Valles Caldera National Preserve Jemez Springs NM USA [Purchase Order NoVCT0207]

Brown D E (1994) Biotic communities Southwestern United States and Northwestern Mexico Lake City UT University of Utah Press

Collins S L amp Calabrese L B (2012) Effects of fire grazing and topographic variation on vegetation structure in tallgrass prai-rie Journal of Vegetation Science 23 563ndash575 httpsdoiorg101111j1654-1103201101369x

CollinsSLampGlennSM(1991)Importanceofspatialandtemporaldynamics in species regional abundance and distribution Ecology 72 654ndash664 httpsdoiorg1023072937205

Collins S L amp Smith M D (2006) Scale- dependent inter-action of fire and grazing on community heterogene-ity in tallgrass prairie Ecology 87 2058ndash2067 httpsdoiorg1018900012-9658(2006)87[2058SIOFAG]20CO2

Conver J L Falk D A Yool S R amp Parmenter R R (2018) Modeling firepathwaysinmontanegrasslandminusforestecotonesFire Ecology 14 17ndash31

Coop J D amp Givnish T J (2007a) Spatial and temporal patterns of recent forest encroachment in montane grasslands of the Valles Caldera New Mexico USA Journal of Biogeography 34 914ndash927 httpsdoiorg101111j1365-2699200601660x

DahmCNCandelaria-LeyRIRealeCSRealeJKampVanHornD J (2015) Extreme water quality degradation following a cata-strophic forest fire Freshwater Biology 60 2584ndash2599 httpsdoiorg101111fwb12548

Debenedetti S H amp Parsons D J (1984) Postfire succession in a Sierran subalpine meadow The American Midland Naturalist 111 118ndash125 httpsdoiorg1023072425549

Dewar J J (2011) Fire history of montane grasslands and ecotones of the Valles Caldera New Mexico USA Thesis University of Arizona

Facelli J M amp Pickett S T A (1991) Plant litter Its dynamics andeffects on plant community structure Botanical Review 57 1ndash32 httpsdoiorg101007BF02858763

Ford P L amp McPherson G R (1996) Ecology of fire in shortgrass prairie ofthesouthernGreatPlainsInDMFinch(Ed)Ecosystem disturbance and wildlife conservation in western grasslands - A symposium proceed-ings (pp 20ndash39) September 22-26 1994 Albuquerque NM General TechnicalReportRM-GTR-285FortCollinsCOUSDAForestServiceRocky Mountain Forest and Range Experiment Station

FosterBLampGrossKL(1998)SpeciesrichnessinasuccessfulgrasslandEffects of nitrogen enrichment and plant litter Ecology 79 2593ndash2602 httpsdoiorg1018900012-9658(1998)079[2593SRIASG]20CO2

Goff F (2009) Valles Caldera A geologic history Albuquerque NM University of New Mexico Press

Hunter M E Omi P N Martinson E J amp Chong G W (2006)Establishment of non- native plant species after wildfires Effects of fuel treatments abiotic and biotic factors and postfire grass seed-ing treatments International Journal of Wildland Fire 15 271ndash281 httpsdoiorg101071WF05074

Keeley JEPausas JGRundelPWBondW JampBradstockRA (2011) Fire as an evolutionary pressure shaping plant traits Trends in Plant Science 16 406ndash411 httpsdoiorg101016jtplants201104002

Milchunas D G amp Lauenroth W K (1993) Quantitative effectsof grazing on vegetation and soils over a global range of en-vironments Ecological Monographs 63 327ndash366 httpsdoiorg1023072937150

Mueggler W F amp Stewart W L (1980) Grassland and shrubland habi-tat types of western Montana US Department of Agriculture Forest ServiceIntermountainForestandRangeExperimentStation[GenTechRepINT-66]OgdenUT

Muldavin E (2003) A vegetation survey and preliminary ecological assess-ment of Valles Caldera National Preserve New Mexico New Mexico


SUAZO et Al

Natural Heritage Program University of New Mexico Albuquerque NM [Cooperative Agreement No 01CRAG0014]

National Cooperative Soil Survey Natural Resources Conservation Service United States Department of Agriculture (1987) Soil survey of Sandoval County Area New Mexico

OesterheldMLoretiJSemmartinMampParueloJM(1999)Grazingfire and Climate effects on primary productivity of Grasslands and Savanna In LRWalker (Ed)Ecosystems of disturbed grounds (pp 287ndash306) Series Ecosystems of the World New York NY Elsevier

Parmenter R (2008) Long- Term Effects of a Summer Fire on Desert Grassland Plant Demographics in New Mexico Rangeland Ecology Management 61 156ndash168 httpsdoiorg10211107-0101

Pickett S T A amp White P S (1985) The ecology of natural disturbance and patch dynamics New York NY Academic Press

RipleyBDonaldGOsborneCPAbrahamTampMartinT (2010)Experimental investigation of fire ecology in the C3 and C4 sub-species of Alloteropsis semialata Journal of Ecology 98 1196ndash1203 httpsdoiorg101111j1365-2745201001700x

Scheintaub M R Derner J D Kelly E F amp Knapp A K (2009)Response of the shortgrass steppe plant community to fire Journal of Arid Environments 73 1136ndash1143 httpsdoiorg101016jjaridenv200905011

ScheiterSHigginsSIOsborneCPBradshawCLuntDRipleyB S hellip Beerling D J (2012) Fire and fire- adapted vegetation pro-moted C4 expansion in the late Miocene New Phytologist 195 653ndash666 httpsdoiorg101111j1469-8137201204202x

Tilman D amp Downing J A (1994) Biodiversity and stability in grass-lands Nature 367 363ndash365 httpsdoiorg101038367363a0

TixDampCharvatI(2005)Abovegroundbiomassremovalbyburningand raking increases diversity in a reconstructed prairie Restoration Ecology 13 228ndash238

VanAuken O W (2009) Causes and consequences of woody plantencroachment into western North American grasslands Journal of Environmental Management 90 2931ndash2942 httpsdoiorg101016jjenvman200904023

Vermeire L T Crowder J L amp Wester D B (2014) Semiarid rangeland is resilient to summer fire and postfire grazing utilization Rangeland Ecology amp Management 67 52ndash60 httpsdoiorg102111REM-D-13-000071

White P S amp Jentsch A (2001) The search for generality in studies of disturbance and ecosystem dynamics Progress in Botany 62 399ndash499

Wright H A amp Bailey A W (1980) Fire ecology and prescribed burning in the Great Plains a research reviewGeneralTechnicalReportINT-GTR-77OgdenUTUSDAForestServiceIntermountainForestandRange Experiment Station p 60

Xiong S amp Nilsson C (1999) The effects of plant litter on vegeta-tion A meta- analysis Journal of Ecology 87 984ndash999 httpsdoiorg101046j1365-2745199900414x

York A M Shrestha M Boone C G Zhang S Harrington J A Jr Prebyl T J hellip Skaggs R (2011) Land fragmentation under rapid urbanization A cross- site analysis of Southwestern cit-ies Urban Ecosystems 14 429ndash455 httpsdoiorg101007s11252-011-0157-8

SUPPORTING INFORMATION

Additional supporting information may be found online in the SupportingInformationsectionattheendofthearticle

Appendix S1 Distribution of Mountain Valley grassland study sites in the Valles Caldera National Preserve New Mexico with Las Conchas Fire burn areaAppendix S2 Plant Species Abundance Data All species averages per treatment group for each year no transformationAppendix S3 Plant species characteristicsAppendix S4 (A) Average precipitation accumulation for June and July averaged over four weather stations (B) Precipitation accumu-lationbywateryear(OctndashSept)averagedoverfourweatherstations

How to cite this article Suazo MM Collins SL Parmenter RR Muldavin E Montane valley grasslands are highly resistant to summer wildfire J Veg Sci 2018291017ndash1028 httpsdoiorg101111jvs12690


SUAZO et Al

Coop amp Givnish 2007a VanAuken 2009) have all contributed to the decline of grassland ecosystems in western North America Fire is an essential ecological component for maintaining the health and proper functioning of forests across the southwestern United States by maintaining community structure ecosystem functioning and species diversity via fuel removal and the release of organically stored nutrients (Dewar 2011) Because montane grasslands are in-terspersed throughout mountain ranges within forest stands there can be important fire regime interactions at a landscape scale that have implications for impacts in more localized habitats For exam-ple in montane grasslands of the Jemez Mountains in north- central New Mexico frequent fires (fire return intervals of lt 10 year) typ-ically move along forestndashgrassland ecotones killing seedling trees encroaching into grasslands (Conver Falk Yool amp Parmenter 2018) Fire has been the single most important factor in preventing forest encroachment and maintaining the integrity of grassland communi-ties however as a result of anthropogenic- induced fire suppression nearly 18 of grassland in this area has been lost to tree encroach-ment between 1935 and 1996 (Coop amp Givnish 2007a)

Most grassland studies of fire impacts focus on warm- season C4- dominated systems at relatively lower elevation with lower av-erage annual precipitation Studies have found that fire favours C4 grasses (Collins amp Calabrese 2012 Tix amp Charvat 2005) and that the expansion of C4 grasses in the late Miocene was driven by frequent fire (Scheiter et al 2012) Therefore it is generally thought that C4 species are better adapted to fire than cool- season C3 species (RipleyDonaldOsborneAbrahamampMartin2010)Consequentlymost studies on fire impacts to grasslands are conducted in C4- dominated grasslands thereby leaving the more mesic C3 grassland response to fire less well documented Also lacking have been the long- term studies of wildfire impacts that include knowledge of pre- fire conditions and postfire succession beyond the first two or three growing seasons following fire Furthermore the effects of fire on species in mountain grasslands are often studied in forest understo-ries rather than open grassland ecosystems

In this context we investigated the effects of an extensivesummer wildfire on plant community structure in the primarily C3- dominated montane valley grasslands of the Valles Caldera National Preserve located in the centre of the Jemez Mountains in north-ern New Mexico USA Using an established long- term pre- and post- wildfire grassland sampling array we asked (a) how did these montane grasslands respond in terms of plant species composition abundance and spatial distribution before and after the wildfire and (b) what biotic and abiotic factors are most influential in stabilizing andor driving change before and after burning

2emsp |emspMETHODS

21emsp|emspSite description

The Valles Caldera National Preserve (Preserve) is located in the Jemez Mountains of north- central New Mexico USA (Supporting

InformationAppendixS1)andconsistsofhigh-elevationforestandgrassland systems that lie at the head of the Jemez River watershed which flows into the Rio Grande Valley The 36017 ha Preserve is centred on a volcanic caldera approximately 24 km in diameter that erupted 12 million years ago (Goff 2009) and comprises large areas of low- lying mountain meadows and riparian vegetation with sloped mountain valley grasslands bordered by several forested lava domes ranging in elevation from 2590 to 3505 m An estimated 10522 ha of montane grassland occurs in the Preserve dominated mostly by C3 grasses that are typical of Southern Rocky Mountain MontanendashSubalpine Grassland per the US National Classification (httpusnvcorg) These include perennial bunch grasses such as Festuca idahoensis F arizonica F thurberi Danthonia parryi and Poa prat-ensis along with a relatively cold- tolerant C4 species Muhlenbergia montana which commonly occurs in subalpine grasslands across the southwestern US (Brown 1994) The grasslands of the Preserve have formed on prehistoric lake beds that are between 70000 and 500000+ years old whose soils consist primarily of Mollisols that have developed from older quaternary alluvial fan (Qf) deposits from the surrounding domes (Muldavin 2003)

Prior to 1900 fires occurred on average every 16 year some-where on Preserve grasslands with widespread fires occurring at decadal intervals and smaller low intensity fires particularly in valley grasslandsoccurringevery27ndash101year(Dewar2011)Inadditionthe Preserve has undergone extensive livestock grazing during the 19th and 20th centuries but currently supports only a relatively small (lt800 head) herd of cattle and a population of approximately 2000 elk (Cervus canadensis) native to the region (cattle have grazed all study sites to a limited degree both pre- and postfire) Mean an-nualprecipitationderived from the2004ndash2015wateryears (OctndashSept) is 545 mm 60 of this precipitation is produced by monsoonal rainfall between the months of June and September and winter pre-cipitation is primarily in the form of snowpack (data from Western Regional Climate Center httpswrccdrieduvallescaldera)

In summer 2011 the Las Conchas Wildfire burned through63131 ha of forest and grassland vegetation and at the time was the largest wildfire in recorded New Mexico history Fire intensity varied from stand replacement in the forest to low- intensity ground fire in the open grasslands The grassland sites in this study burned over the course of 5 days from 27 June to 1 July with an average wind speed of 2 ms and gusts with a maximum of 171 ms There was 36 average relative humidity with a 103 minimum and 892 maximum Average air temperature for this time period was 180degC 36degC minimum and 283degC maximum

22emsp|emspField data collection

A rangeland monitoring programme was established in the Preserve in 2001 consisting of 44 vegetation monitoring sites (Barnes 2003) These sites were stratified by relatively homogenous repeating ecotypes which were identified by soil type land form community composition management history and future management poten-tial Three ecotypes occur across the Preserve Mountain Meadow


SUAZO et Al








(a) (d)

(b) (e)

(c) (f)


SUAZO et Al

TABLE 1emsp

Site

cha

ract

eris

tics

for a

ll m

onito

ring

plot

s

Site

Trea

tmen

tTo

tal y

ears

gr

azed

Year

s gra

zed

pre-

fire

Year

s gra

zed

post

- fire

Year

s NO

T sa

mpl

edSo

il se

riesa

Asp

ect (

deg)Sl

ope

El

evat

ion

(m)

Dom

inan

t gra

ssb

MV

02BU

RN6

60

130

221

12

260

7FESX1MUMOPOPR

MV

03BU

RN6

60

330

818

15

261

1FESX1MUMOPOPR

MV

04BU

RN7

34

130

215

34

262

3DAPAPOPRFESX1

MV

05BU

RN6

24

230

468

52

734

DAPAPOPRFESX1

MV

06BU

RN5

50

130

245

32

656

DA

PA F

ESX1

MV

07BU

RN6

60

530

432

88

269

2D

APA

FES

X1

MV

08BU

RN5

50

130

419

88

270

2D

APA

FES

X1

MV

09CONTROL

66

01

302

168

22

585

FESX1MUMO

MV1

0CONTROL

66

02

302

245

42

559

FESX1MUMO

MV1

1CONTROL

116

53

302

269

112

574

FESX1POPR

MV1

2CONTROL

116

52

304

626

268

5DAPAPOPRFESX1

MV1

3CONTROL

116

53

304

274

266

2DAPAPOPRFESX1

MV1

5BU

RN12

75

430

467

82

650

DAPAPOPRFESX1

MV1

7CONTROL

138

52

302

123

32

608

POPR

MV1

8BU

RN12

75

530

250

72

670

DAPAPOPRFESX1

MV

20CONTROL

116

54

304

365

261

4DAPAPOPR

a Soil

Serie

s ta

ken

from

USD

A N

RCS

Soil

Surv

ey o

f San

dova

l Cou

nty

Are

a N

ew M

exic

o b D

omin

ant p

lant

spe

cies

by

abun

danc

e D

APA

Dan

thon

ia p

arry

i FE

SX1

Fes

tuca

ariz

onic

a an

d Fe

stuc

a id

ahoe

nsis

consolidatedMUMOM

uhle

nber

gia

mon

tanaPOPRP

oa p

rate

nsis


SUAZO et Al








3emsp |emspRESULTS




SUAZO et Al






PERMANOVA PERDISPa





9 7 12 01 963 0 08 999



Pre- fire AvAbund



Agrostis scabra 12 0 112 157 323 747


Poa fendleriana 157 085 108 133 311 137




Pre(2010)- Post(2012)

PERMANOVA PERDISPa





7 7 07 09 744 13 02 999



Pre- fire AvAbund





Agrostis scabra 12 034 102 141 289 1382



SUAZO et Al







PERMANOVA PERDISPa





40 28 16 001 999 2 0009 999



Pre- fire AvAbund




Poa fendleriana 143 092 099 128 272 976





Pre- fire AvAbund




Poa fendleriana 143 108 096 123 259 96










Poa fendleriana 092 108 089 117 24 1181




SUAZO et Al













SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al








(a) (d)

(b) (e)

(c) (f)


SUAZO et Al

TABLE 1emsp

Site

cha

ract

eris

tics

for a

ll m

onito

ring

plot

s

Site

Trea

tmen

tTo

tal y

ears

gr

azed

Year

s gra

zed

pre-

fire

Year

s gra

zed

post

- fire

Year

s NO

T sa

mpl

edSo

il se

riesa

Asp

ect (

deg)Sl

ope

El

evat

ion

(m)

Dom

inan

t gra

ssb

MV

02BU

RN6

60

130

221

12

260

7FESX1MUMOPOPR

MV

03BU

RN6

60

330

818

15

261

1FESX1MUMOPOPR

MV

04BU

RN7

34

130

215

34

262

3DAPAPOPRFESX1

MV

05BU

RN6

24

230

468

52

734

DAPAPOPRFESX1

MV

06BU

RN5

50

130

245

32

656

DA

PA F

ESX1

MV

07BU

RN6

60

530

432

88

269

2D

APA

FES

X1

MV

08BU

RN5

50

130

419

88

270

2D

APA

FES

X1

MV

09CONTROL

66

01

302

168

22

585

FESX1MUMO

MV1

0CONTROL

66

02

302

245

42

559

FESX1MUMO

MV1

1CONTROL

116

53

302

269

112

574

FESX1POPR

MV1

2CONTROL

116

52

304

626

268

5DAPAPOPRFESX1

MV1

3CONTROL

116

53

304

274

266

2DAPAPOPRFESX1

MV1

5BU

RN12

75

430

467

82

650

DAPAPOPRFESX1

MV1

7CONTROL

138

52

302

123

32

608

POPR

MV1

8BU

RN12

75

530

250

72

670

DAPAPOPRFESX1

MV

20CONTROL

116

54

304

365

261

4DAPAPOPR

a Soil

Serie

s ta

ken

from

USD

A N

RCS

Soil

Surv

ey o

f San

dova

l Cou

nty

Are

a N

ew M

exic

o b D

omin

ant p

lant

spe

cies

by

abun

danc

e D

APA

Dan

thon

ia p

arry

i FE

SX1

Fes

tuca

ariz

onic

a an

d Fe

stuc

a id

ahoe

nsis

consolidatedMUMOM

uhle

nber

gia

mon

tanaPOPRP

oa p

rate

nsis


SUAZO et Al








3emsp |emspRESULTS




SUAZO et Al






PERMANOVA PERDISPa





9 7 12 01 963 0 08 999



Pre- fire AvAbund



Agrostis scabra 12 0 112 157 323 747


Poa fendleriana 157 085 108 133 311 137




Pre(2010)- Post(2012)

PERMANOVA PERDISPa





7 7 07 09 744 13 02 999



Pre- fire AvAbund





Agrostis scabra 12 034 102 141 289 1382



SUAZO et Al







PERMANOVA PERDISPa





40 28 16 001 999 2 0009 999



Pre- fire AvAbund




Poa fendleriana 143 092 099 128 272 976





Pre- fire AvAbund




Poa fendleriana 143 108 096 123 259 96










Poa fendleriana 092 108 089 117 24 1181




SUAZO et Al













SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al

TABLE 1emsp

Site

cha

ract

eris

tics

for a

ll m

onito

ring

plot

s

Site

Trea

tmen

tTo

tal y

ears

gr

azed

Year

s gra

zed

pre-

fire

Year

s gra

zed

post

- fire

Year

s NO

T sa

mpl

edSo

il se

riesa

Asp

ect (

deg)Sl

ope

El

evat

ion

(m)

Dom

inan

t gra

ssb

MV

02BU

RN6

60

130

221

12

260

7FESX1MUMOPOPR

MV

03BU

RN6

60

330

818

15

261

1FESX1MUMOPOPR

MV

04BU

RN7

34

130

215

34

262

3DAPAPOPRFESX1

MV

05BU

RN6

24

230

468

52

734

DAPAPOPRFESX1

MV

06BU

RN5

50

130

245

32

656

DA

PA F

ESX1

MV

07BU

RN6

60

530

432

88

269

2D

APA

FES

X1

MV

08BU

RN5

50

130

419

88

270

2D

APA

FES

X1

MV

09CONTROL

66

01

302

168

22

585

FESX1MUMO

MV1

0CONTROL

66

02

302

245

42

559

FESX1MUMO

MV1

1CONTROL

116

53

302

269

112

574

FESX1POPR

MV1

2CONTROL

116

52

304

626

268

5DAPAPOPRFESX1

MV1

3CONTROL

116

53

304

274

266

2DAPAPOPRFESX1

MV1

5BU

RN12

75

430

467

82

650

DAPAPOPRFESX1

MV1

7CONTROL

138

52

302

123

32

608

POPR

MV1

8BU

RN12

75

530

250

72

670

DAPAPOPRFESX1

MV

20CONTROL

116

54

304

365

261

4DAPAPOPR

a Soil

Serie

s ta

ken

from

USD

A N

RCS

Soil

Surv

ey o

f San

dova

l Cou

nty

Are

a N

ew M

exic

o b D

omin

ant p

lant

spe

cies

by

abun

danc

e D

APA

Dan

thon

ia p

arry

i FE

SX1

Fes

tuca

ariz

onic

a an

d Fe

stuc

a id

ahoe

nsis

consolidatedMUMOM

uhle

nber

gia

mon

tanaPOPRP

oa p

rate

nsis


SUAZO et Al








3emsp |emspRESULTS




SUAZO et Al






PERMANOVA PERDISPa





9 7 12 01 963 0 08 999



Pre- fire AvAbund



Agrostis scabra 12 0 112 157 323 747


Poa fendleriana 157 085 108 133 311 137




Pre(2010)- Post(2012)

PERMANOVA PERDISPa





7 7 07 09 744 13 02 999



Pre- fire AvAbund





Agrostis scabra 12 034 102 141 289 1382



SUAZO et Al







PERMANOVA PERDISPa





40 28 16 001 999 2 0009 999



Pre- fire AvAbund




Poa fendleriana 143 092 099 128 272 976





Pre- fire AvAbund




Poa fendleriana 143 108 096 123 259 96










Poa fendleriana 092 108 089 117 24 1181




SUAZO et Al













SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al








3emsp |emspRESULTS




SUAZO et Al






PERMANOVA PERDISPa





9 7 12 01 963 0 08 999



Pre- fire AvAbund



Agrostis scabra 12 0 112 157 323 747


Poa fendleriana 157 085 108 133 311 137




Pre(2010)- Post(2012)

PERMANOVA PERDISPa





7 7 07 09 744 13 02 999



Pre- fire AvAbund





Agrostis scabra 12 034 102 141 289 1382



SUAZO et Al







PERMANOVA PERDISPa





40 28 16 001 999 2 0009 999



Pre- fire AvAbund




Poa fendleriana 143 092 099 128 272 976





Pre- fire AvAbund




Poa fendleriana 143 108 096 123 259 96










Poa fendleriana 092 108 089 117 24 1181




SUAZO et Al













SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al






PERMANOVA PERDISPa





9 7 12 01 963 0 08 999



Pre- fire AvAbund



Agrostis scabra 12 0 112 157 323 747


Poa fendleriana 157 085 108 133 311 137




Pre(2010)- Post(2012)

PERMANOVA PERDISPa





7 7 07 09 744 13 02 999



Pre- fire AvAbund





Agrostis scabra 12 034 102 141 289 1382



SUAZO et Al







PERMANOVA PERDISPa





40 28 16 001 999 2 0009 999



Pre- fire AvAbund




Poa fendleriana 143 092 099 128 272 976





Pre- fire AvAbund




Poa fendleriana 143 108 096 123 259 96










Poa fendleriana 092 108 089 117 24 1181




SUAZO et Al













SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al







PERMANOVA PERDISPa





40 28 16 001 999 2 0009 999



Pre- fire AvAbund




Poa fendleriana 143 092 099 128 272 976





Pre- fire AvAbund




Poa fendleriana 143 108 096 123 259 96










Poa fendleriana 092 108 089 117 24 1181




SUAZO et Al













SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al













SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al








Richness 1 18 0611 043

Evenness 1 10376 0435 051

Grass 1 133 0018 089

Forb 1 4025 0819 037


Plant height 1 3315 47563 003

Richness 1 709 30605 008

Evenness 1 402486 146929 lt005

Grass 1 51215 78948 lt005

Forb 1 50697 94613 lt005

p-value 005 = 001 = 0001 =



SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al










SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al



ACKNOWLEDG EMENTS


ORCID


R E FE R E N C E S






















SUAZO et Al






















SUAZO et Al





















montane valley grasslands are highly resistant to summer...

Documents